Soil Hydraulic Functions Revisited: Challenges and Solutions.

The conceptual framework of the Richards equation with parametric soil hydraulic functions (SHP) as constitutive relationships is the standard approach to describe transient water transport in unsaturated soils, but also heavily criticized as being inadequate. Based on the seminal work of Gardner, Brooks and Corey, van Genuchten and others, there is grown knowledge about suitable parameterizations of SHP, but a thorough examination shows that the use of these traditional functions is not as successful as we like to believe. Expressed provocatively, great challenges are found (i) at the wet end of the moisture range, (ii) at the dry end of the moisture range, and (iii) in the whole range in between. Problems refer to both the retention function and in particular to the shape of the conductivity function that is derived from the soil water retention curve by capillary bundle models.

Typical questions to pose are: What is the right degree of flexibility in describing SHP? How do we handle the properties near the wet end, in particular if the conductivity curve is predicted from the retention curve? What is an appropriate model of SHP in situations of evaporation, when soil gets dry? And do we have to care for the fact that SHP under transient conditions may differ from SHP obtained under equilibrium measurements, caused by dynamic non-equilibrium (DNE) and hysteresis?

Our presentation shows that for many of the problems there are good solutions available, but they are rarely applied. For challenges (i) and (ii), models are available that are structurally more adequate than the traditional approaches, with no need for additional parameters. For solutions with respect to (iii), the price to pay is an increased number of parameters, which inevitably requires an increased effort for parameter identification,   i.e., better experiments and evaluation methods.